QiLife User Flows

1. Quick Capture and AI Triage

sequenceDiagram
    actor Cody
    participant UI as QiLife Web UI
    participant API as FastAPI Backend
    participant AI as AI Service
    participant DB as SQLite Database

    Cody->>UI: Enter raw text in Quick Capture
    UI->>API: POST /api/qibits/capture
    API->>DB: Save QiBit with status 'new'
    API->>AI: interpret_qibit(raw_capture)
    AI-->>API: Suggested structure
    API->>DB: Save suggestions in ai_outputs
    API-->>UI: Return QiBit plus staged suggestions
    UI-->>Cody: Show QiBit in Inbox with review controls
    Cody->>UI: Accept or edit suggestions
    UI->>API: POST /api/qibits/{id}/triage
    API->>DB: Update QiBit and create approved downstream records
    API->>DB: Write provenance links
    UI-->>Cody: Remove from Inbox and expose in Timeline/Thread/Action views

2. Daily Planning

1. Cody opens Today.
2. QiLife shows:
3. due or scheduled actions
4. open loops (waiting_on actions and unresolved obligations)
5. recent QiBits
6. an AI focus summary placeholder or live summary
7. Cody schedules or reschedules actions by writing scheduled_for.
8. If an item is blocked, Cody sets the action or thread status to waiting_on.

3. Active Work with Context Dock

1. Cody opens a thread, action, QiBit, or person detail view.
2. The Context Dock shows:
3. linked knowledge items
4. linked documents
5. related QiBits
6. people/entities
7. obligations and transactions
8. prior resolutions and recent activity
9. Cody completes the work in the center pane while the Context Dock keeps the relevant knowledge beside it.

4. End-of-Day Review

1. Cody opens the end-of-day review in Today.
2. The backend aggregates the day's projected timeline items.
3. summarize_day() generates or refreshes a daily_summaries record for the date.
4. generate_reflection_prompt() suggests a reflection prompt.
5. Cody may save a reflection as a QiBit of type reflection.

daily_summaries and reflection QiBits are distinct:

• daily_summaries are synthesized day-level summaries
• reflection QiBits are authored reflections tied to lived events

5. Ask QiLife

1. Cody asks a question in Ask QiLife.
2. The backend searches links, thread context, bucket context, SQL matches, and text search.
3. The answer returns cited supporting records.
4. Cody can save the output as:
5. action
6. knowledge item
7. note-type QiBit
8. reflection-type QiBit